This paper presents a multiphysical battery pack model, along with a procedure to identify its parameters and its application to state-of-charge (SOC) determination using an extended Kalman filter (EKF). The model enables the reproduction of the electric and thermal behaviors of batteries with high accuracy. A methodology is proposed to identify the parameters of the model and optimize them. The model is experimentally validated with measurements run on high-energy-density ThunderSky cells. A comparison of measurements and model results shows that the electrical model error is below 1.6% and that the error of the thermal model is below 2.5%. The model is used as an EKF basis to reliably estimate the battery SOC. The results are also experimentally verified through measurements showing that the proposed model performs better than other simpler models and determination methods.Index Terms-Kalman filter, lithium-ion battery, multiphysical modeling, parameter identification, state-of-charge (SOC) determination.